← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6650.g · 生成模型:GPT-6

误报与漏报可在同一规则共存

理解等级:中级

模式太宽会把不应识别的文本判为目标,太窄会漏掉真实目标。书中复杂问候规则既漏掉hi Rosa,又接收ello Rosa,说明一条规则可同时存在两种错误;增加规则前应准备真实正反样本并明确目标定义。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      There is a lot of logic packed into that first line of code, the regular expression, and it gets the job done for a surprising range of greetings. But it missed that Manning typo, which is one of the reasons NLP is hard. In machine learning and medical diagnostic testing, that’s called a false negative classification error. Unfortunately, it will also match some statements that

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第01章《会读写的机器与自然语言处理》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-1.html) - 结论依据:摘录中关于误报与漏报可在同一规则共存的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:There is a lot of logic packed into that first line of code, the regular expression, and it gets the job done for a surprising range of greetings. But it missed that Manning typo, which is one of the reasons NLP is hard. In machine learning and medical diagnostic testing, that’s called a false negative classification error. Unfortunately, it will also match some statements that